Abstract Body: Background: Terminal digit preference (TDP) is an overrepresentation of blood pressure (BP) measurements ending in certain end digits, most often 0 or 5, which may result in misclassification of hypertension status and suboptimal treatment decisions. We evaluated whether use of oscillometric devices and BP measurement training for healthcare professionals impacted TDP in primary care settings.
Methods: We analyzed data from 14 clinics in Maryland and DC participating in the Home Blood Pressure Telemonitoring LINKED with Community Health Workers to Improve Blood Pressure (LINKED-BP) Program, a randomized clinical trial of a home BP telemonitoring intervention for adults at risk for stage 2 hypertension. Clinic systolic BP (SBP) and diastolic BP (DBP) measurements were recorded before and after clinical training on proper BP measurement techniques, including positioning, cuff selection, measurement protocols, and TDP. BP measurements from 262 participants pre-training and measurements from 143 participants post-training were included. We examined the distributions of SBP and DBP pre- and post-training and calculated the percent of measurements ending in each end digit 0-9. We used chi-square tests to compare the observed terminal digit distributions to the expected uniform distribution.
Results: At baseline, there was no visual evidence of TDP in the distributions of SBP or DBP measurements. Pre-training, 10.3% of SBP measurements and 8.8% of DBP measurements ended in 0. There was no evidence of a difference in the observed terminal digit distributions for SBP (p=0.978, Figure 1) or DBP (p=0.749) compared to the expected uniform distribution. After BP measurement training, 9.8% of SBP and 8.4% of DBP measurements ended in 0 and there was no evidence of TDP for either SBP (p=0.993, Figure 2) or DBP (p=0.951).
Conclusions: We found no evidence of TDP at pre- or post-training. The results suggest that participating clinics were already employing appropriate BP measurement recording practices, possibly due to annual education, prior participation in research, and widespread use of oscillometric devices. These findings highlight the importance of evaluating BP measurement quality in hypertension prevention and management studies to tailor BP measurement training. Future research could explore the impact of training on other aspects of the BP measurement procedure and factors contributing to high-quality BP measurements in primary care.